Maternal Grandsire

The Maternal Grandsire
By Cindy Vogels

Genetics work in horses and mice has produced revolutionary and exciting new insights that may influence your canine breeding decisions.

For years, horsemen have acknowledged a phenomenon called the maternal-grandsire effect, when outstanding males do not immediately reproduce their greatness in the next generation. Instead, they produce daughters who are outstanding dams. An oft-cited example is Secretariat, perhaps the greatest thoroughbred of all time. Secretariat's achievement was not matched by his direct get, who by and large were unremarkable, but rather was passed on through his daughters, many of whom went on to produce great performers. Dog breeders, too, have noted that an extraordinary male, while not producing extraordinary offspring, will often produce daughters who are prolific and exceptional dams. For years, there was absolutely no scientific explanation of this phenomenon in which traits skip a generation and are passed along only by female offspring. Recently, however, an article documenting scientific evidence of the maternal-grandsire effect appeared in issue number 242 of Equus, an outstanding horse publication. I acknowledge that article for providing me with much of the information in this column.

Some Genetics Background
In each cell of a dog's body there are 39 pairs of chromosomes, one set from each parent. Each chromosome pairs off with a corresponding chromosome of the other parent, and in each chromosome there are thousands of genes, which contain the protein codes that determine every physical trait. Within a pair of chromosomes will be pairs of genes from each parent that determine various traits. When the genes are not in conflict with each other - both expressing brown eyes, for example - there is no problem. However, if one chromosome contains the gene for brown eyes but another one contains the genes for green eyes, long-accepted Mendelian theory states that only the genetically dominant chromosome will be expressed. The theory also states that genetic dominance is unrelated to the sex of the gene donor. When both genes are expressed, they are considered to be co-dominant. Coat color, for example, is an area in which both genes can sometimes exert influence. Other times, both genes are recessive, but one is nonetheless more dominant than the other, thus allowing a recessive gene to be expressed. Recessive genes may also be expressed when both contain the same protein code for a trait.

A Startling Study
In 1969, Dr. W.R. Allen startled the world with a study that seemed to indicate certain genes might be gender-related in their expression. Allen bred horses and donkeys, and during pregnancy measured levels of the pregnancy hormone called equine chorionic gonadotrophin (ECG). Normally this level is high in horse-horse crosses and low in donkey-donkey crosses. According to Mendel, it should not have made any difference which species served as sire or dam. The levels should reflect a combination of the two species, and would either be a moderate level (indicating co-dominance), or if one species dominated, the level would be either high or low. Surprisingly, the mares (horse females) bred to donkeys exhibited low levels of ECG, much like a donkey-donkey cross, and the jennies (female donkeys) bred to horses registered high levels of ECG, as in a horse-horse cross. Although no definitive conclusions were reached, it appeared that the sires' genes were the only factor affecting the ECG levels in the females. The females' genes were silent.

It was not until 1986 that the topic reappeared in the literature. A research team headed by Dr. Azim Surani used mice to create embryos in which all the genetic material was received entirely from either one parent or the other. Since the material was transmitted in appropriately matched pairs, Mendelian theory would have predicted that the embryos would develop normally, since it was only the presence of two genes for each trait, and not the sex of the gene donors, that was considered relevant. Again, however, Mendelian expectations were confounded, as the all-female gene pairings resulted in large placentas with little embryonic material. The all-male gene pairings produced the opposite result: small placentas with large embryos. Surani's team concluded that some genes do not follow Mendel's laws. Some are "switched on" before fertilization and are always expressed, while others are "switched off" and never expressed. The sex of the gene donor is the factor that determines which mode a gene will fall into. A theory called "genome imprinting" was created to account for this previously unformulated phenomenon.

For example, say there is a canine gene that is paternally imprinted and, when expressed, produces three-eared dogs. When the gene is not expressed, the dog has two ears. A three-eared male inherits the gene from his mother, but because a gene that is paternally imprinted is switched off when passed on by a male to its offspring, he will have all two-eared offspring. His male two-eared offspring will not produce three-eared dogs, but his daughters will, because a gene that is paternally imprinted will be switched on in females.

Questions and Implications
Many questions still remain, and the literature is vague on why the phenomenon might occur. Researchers point to the significance of gender-related functions. For example, it appears that males strive to produce virulent, robust get, while females, for their own well-being, control the size of their offspring. Imprinted genes are quite possibly involved in traits inherited polygenically. If only some of the genes are switched on, the work of the geneticist tracking inheritance becomes more complicated.

The implications of this finding go far beyond the world of Thoroughbred racers. Already, a number of imprinted human genes have been pinpointed. Ongoing mapping of the canine genome should increase the likelihood of detecting imprinted genes in dogs. The most important contribution would probably be in the realm of canine health, but eventually we might have the tools to track the inheritance of many canine characteristics that seem capricious in their skipping of generations.

Dog breeders should be aware of this possible maternal-grandsire effect. Keep in mind, however, that outstanding males tend to be bred to outstanding females, so even if some of the male's desirable genes are paternally imprinted, the offspring of such matings will probably inherit some excellent traits from their exceptional dams. For example, this year's Kentucky Derby and Preakness winner, Charismatic, was sired by 1990 Preakness winner Summer Squall, who is out of a Secretariat daughter. While Summer Squall's prowess on the track could be traced to the maternal-grandsire effect, he seemed to pass his greatness along directly to Charismatic. However, Secretariat's mother appears another time in Charismatic's pedigree and Secretariat's sire Bold Ruler appears twice. So, the talented colt's lineage points back to many outstanding individuals. A pedigree, whether for dogs or horses, always contains many influences and variables. We dog breeders tend to be impatient and are disappointed when an outstanding male does not immediately reproduce his excellence. Remember the maternal-grandsire effect, and wait a generation.

Cindy Vogels is breeder-judge from Littleton, Colo. She has bred Soft Coated Wheaten Terriers, Kerry Blue Terriers, Welsh Terriers and other breeds for almost 30 years, and judges 18 terrier breeds.

AKC GAZETTE articles are selected for their general interest and entertainment values. Authors' views do not necessarily represent the policies of the American Kennel Club, nor does their publication constitute an endorsement by the AKC.


The Nature of Genetic Disease

The Nature of Genetic Disease
by John Armstrong

Many people label any problem that appears to be inherited a "genetic disease". However, though there are legitimate genetic diseases, there are also a variety of problems that have an inherited component, but are of a fundamentally different nature. Dealing effectively with any genetic problem requires an understanding of the relationship between the genes (genotype) and the phenotype. In many cases this is lacking. In this article, I would like to describe some of the differences in order to give breeders and owners a better understanding of what they are dealing with.

Inborn errors of metabolism: the true genetic "diseases"
The first clearly-described relationship between genotype and metabolic deficiencies is credited to Sir Archibald Garrod, an English physician. In 1901, he showed that the inherited disease alkaptonuria results from an inability to metabolize certain amino acids, leading to the accumulation of homogentisic acid. Some of this compound accumulates in skin and cartilage (the latter leading to arthritis). The rest is excreted in the urine, turning it black. Garrod suggested that the metabolic block was caused by an enzyme deficiency, though this was not confirmed until the enzyme (homogentisic acid oxidase) was characterized in 1958.

Since Garrod's time, many other inherited metabolic diseases have been discovered. Some can be managed by careful attention to diet; others cannot. A particularly nasty example is Tay- Sachs disease, which involves an enzyme important in lipid metabolism. Individuals homozygous for a deficiency in this enzyme accumulate a compound called a ganglioside in the nervous system. They appear normal at birth but progressively lose motor functions and die around 3 years of age. There is no treatment.

Most of these conditions involve mutations that lead to the production of a nonfunctional enzyme, or one that is totally absent. In heterozygotes, the single good copy of the gene is generally able to produce sufficient enzyme to handle the normal workload. However, in a few cases, carriers as well as affected individuals have to be careful about their diet, or may exhibit less severe phenotypic effects.

Example of inherited metabolic diseases in dogs include phosphofructokinase deficiency in Cocker and Springer Spaniels, and pyruvate kinase deficiency in Basenjis.

Not all mutations involve metabolic pathways. Some involve proteins that have structural roles in cells and tissues. Others involve regulatory genes that control the correct sequence of events during development. These may lead to such problems as septal defects in the heart or the failure of the embryonic kidney to develop into the adult form. Nevertheless, all can legitimately be considered genetic diseases as there is a direct one-to-one relationship between a single mutated gene and a particular problem.

Conformational "diseases" - the result of unnatural selection
Problems such as hip dysplasia and bloat clearly have a genetic component, but also an environmental component, and perhaps a behavioral one as well (which may also be partially determined by the genes). Gastric torsion is not a genetic disease, and it would be foolish to think that one can identify a single gene responsible for bloat. One might better compare a bloat attack to a bad case of indigestion in a human. Some people are more prone to such attacks than others, and there may well be an inherited component, but other factors play a substantial role. Research into bloat suggests that diet, behavior and conformation may all play a role.

Leaving aside the question of the role of genetics in behavior, the results suggest that the incidence of bloat increases with the size of the dog and the depth and width of the chest cavity. This is a conformational problem, not a genetic disease. Certainly the overall conformation is, ultimately, determined by the genes, but not by a single gene. There is no bloat gene we can identify and eliminate. There are probably dozens or hundreds of genes that go into determining the shape and size of the head, trunk and limbs. Wherever there is genetic variability, one can select for larger, smaller, narrower, wider, etc. Because the fancy as a whole decided that a taller, narrower Standard Poodle looked more "refined", more of that description were kept for breeding purposes and the population has been shifted toward a more bloat-prone conformation. [This is not exclusively a poodle problem. There are other large breeds in which it is even more frequent.]

When it comes to the question of correcting it, the solution, in theory, is simple. We stop breeding for a bloat-prone conformation and go back to a slightly smaller Standard with a chest cavity that is not so deep or narrow. Some may regard this as a retrogressive step, but we have to decide which we want to sacrifice.

I do not rule out the possibility that two dogs of identical conformation may have one or more genes that lead to one being more bloat-prone than the other. If we could identify these genes, we might be able to reduce the incidence somewhat while retaining some of the desired "refinement".

This site is the property of RooBear Yorkies. Please do not steal our pictures